Multiscale metapopulations

Abstract New advents in technology have improved our ability to study and track host-associated organisms. Often, these organisms experience vastly different spatial and temporal scales than their hosts. However, host-associated organisms are not independent of their hosts nor are hosts independent of their host-associated organisms. Rather, processes at the host and host-associated organism scales interact to alter outcomes for both the host and its symbionts. Often, host-associated organisms are described in terms of metapopulations, wherein individual hosts act as habitat patches for their symbionts. When hosts themselves also exist in patchy environments, the result is metapopulation structure at two distinct scales. In this paper, we develop a model to explore how the incorporation of multiscale metapopulation dynamics impacts metapopulation predictions at both the host and symbiont scales. While related to several disease models, our system differs in its inclusion of full colonization and extinction dynamics at both scales, and its emphasis on how the interaction between colonization and extinction at these two scales impacts the host and its host-associated organisms, including pathogens as well as symbionts that are beneficial to the host. Overall, our multiscale metapopulation model predicts rich dynamics that often deviate from the dynamics of metapopulation models built for either the host or its host-associated organisms independently.